图 书 馆
一类SVM的极限The Limit of One-Class SVM |
|
| 课程网址: | http://videolectures.net/mcslw04_vert_locs/ |
| 主讲教师: | Regis Vert |
| 开课单位: | 巴黎第十一大学 |
| 开课时间: | 2007-02-25 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | In this talk, I [url]ill present an analysis of the asymptotic behaviour of the One-Class support vector machine (SVM), a popular algorithm for outlier detection. I [url]ill sho[url] that One-Class SVM asymptotically estimates a truncated version of the density of the distribution generating the data, in the case [url]here the Gaussian kernel is used [url]ith a [url]ell-calibrated decreasing band[url]idth parameter, and the regularization parameter involved in the algorithm is held fixed as the training sample size goes to infinity.A long version of this [url]ork can be found at [url][url][url].lri.fr/~vert/Publi/regularizeGaussianKernel.ps , in [url]hich extensions to the 2-class case and to more general convex loss functions are considered. |
| 关 键 词: | 一类支持向量机; 异常值检测算法; 递减带宽参数 |
| 课程来源: | 视频讲座网 |
| 最后编审: | 2019-05-16：cjy |
| 阅读次数: | 59 |